Title: Investment risk early warning method of listed companies based on EMD-RF-LSTM

Authors: Lu Ren; Wenyu Zhang

Addresses: China Aerospace Academy of Systems Science and Engineering, Beijing, 100035, China; School of Economics and Management, Xi'an University of Posts and Telecommunications, Xi'an, 710061, China ' China Aerospace Academy of Systems Science and Engineering, Beijing, 100035, China; School of Economics and Management, Xi'an University of Posts and Telecommunications, Xi'an, 710061, China

Abstract: In order to overcome the problems of low accuracy and poor correlation in the early warning method of investment risk of listed companies, this paper designs an early warning method of investment risk of listed companies based on EMD-RF-LSTM. First of all, determine the internal and external early warning factors of investment risk, and realise the screening of risk early warning factors. Then, the maximum likelihood estimate of the influencing factors is calculated by clustering algorithm to realise the factor quantification. Finally, the EMD method is used to extract the characteristics of early warning factors, build the early warning function through RF, and solve the critical value of risk early warning through LSTM to achieve risk early warning. The results show that the accuracy of the proposed method can reach 99%, and the correlation coefficient is 0.97, which plays an important role in the early warning of investment risk.

Keywords: EMD method; LSTM method; RF method; clustering algorithm; investment risk warning.

DOI: 10.1504/IJSD.2024.136621

International Journal of Sustainable Development, 2024 Vol.27 No.1/2, pp.170 - 185

Received: 24 Feb 2023
Accepted: 30 Aug 2023
Published online: 08 Feb 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article